Last Saturday, September 2, we hosted a chat in the Halls of Volgaard Discord server. This chat was open to Volland subscribers. This informal chat and Q&A allowed members to chat about how their week went and ask Wizard of Ops any questions. This transcript removes identifying details about chat participants, as well as non-trading related chatter.
Question: At some point during the last one hour on Friday, September 1, aggregate charm was about -5 billion. You indicated that 1 billion is about ten points on a graph. Should one have expected a large move at the end? We know it did not materialize.
Answer: So we're going to get into the weeds of calculation a little bit here. From a natural calculation perspective, charm is supposed to be the movement of Delta for one day passing (like plus one day passing). We have shortened it. We have changed that to one hour of passing. So when you look at charm here in SPX, you are going to see how much they would have to hedge for one hour of passing. When you see 5 billion, you would have to divide it technically by the portion of an hour that you have. We tried to put that into Volland. We tried to, I guess you could say, create a situation where it's always giving you the time until the end of the day. The problem is, when we do that calculation, then it weighs too heavily in the beginning of the day and it weighs things that have like 20 days left due to IV. So we decided to keep it like this - at one hour. So when you see something like 5 billion with like 5 minutes left, you have to divide that 5 billion by 20 in order to get a decent, accurate charm hedging number.
Question: And that relationship is still valid over the last year where options volume has become much shorter dated?
Answer: Are you talking about spot vol correlation relationship? In short, yes, the spot vol correlation actually strengthened recently. A lot of the hedging has been done intraday, like when there's big super moves... intraday vol has changed a bit, but not so much the 30 day vol and market makers still price it. I think it's actually been strengthened by that VIX spot vol correlation.
Question: How do you judge the changes in IV intraday when IV is naturally decaying?
Answer: The natural decay is caught in charm. So, the implied volatility changes are independent of the time aspect. So vanna and charm, while they're related because they're related to the same thing, are measuring the natural decay, which is charm, versus the implied volatility change, which is vanna. And it's easy calculation-wise to separate it, because you already know how much trading time there is to expiration. So you have to wait a little bit, but it's pretty... I don't want to say "simple"... I mean, it's mathematical, but it is a good way to take out fluctuations. You have a standard for how long time takes.
Question: Would gamma have any influence on price action and intraday activity?
Answer: Yes! So the thing with gamma is when you have big strikes in gamma in 0DTE, it's like a little sugar rush... because while the actual number gamma is higher, the I guess you could say "sphere of influence" is a lot narrower. So let's say you're at 4500 or something like that, the gamma exposure is, you know, 2 billion per point, right? Per point of SPX... Let's just say it's like 2 billion or something like that... It is pretty much isolated to that one spot. Like to 4500. I mean, yeah, it expands a little bit to 4510 and a little bit to 4490, but if it's 2 billion a month out at 4500, then it'll be 1 billion out at 4400, it'll be 500 million out at 4350. And typically, the gamma is lower the further out you are. So the sphere of influence is wider, but it's smaller. So when you see something like a 0DTE high gamma strike getting crossed, you might see something like a quick five point jolt, but then it's done very quickly. Because the sphere of influence on that is pretty small. That's because the price changes pretty quickly, too, when you pass those strikes. I like to measure those kinds of things in charm, just because I can conceptually see it that way. And that's why I put charm in the 0DTE alerts. But, you know, gamma has just as much of an impact.
I could tell you what the spot vol correlation is right now... The R squared is 50.57. So yes, it's still strong, but not exceedingly strong. The correlation is -0.75. So basically for every percent gain in SPX, VIX should be coming down by 0.75 points.
Question: Less of a question, more of an observation. I know 0DTE charm works somewhere around 90% for predicting moves. In the off 10%, have you seen a correlation of other market factors which tells you that charm will not work on a certain day?
Answer: There are a couple of things. Number one, the overvixing/undervixing kind of thing means that you have to pay attention to higher-order vanna. Number two, a lot of times we call them "macro bros"... As much as I'd love for option dealers to be the only traders, by definition, the fact that there are people in this room means that option dealers are not the only traders. So, there are other people who are trying to position in the market. There are earnings, there's, you know, all kinds of stuff. So when you say, "What market conditions do I think charm won't work?" I don't think it's necessarily a market condition as much as it is a macro condition or a condition that's outside of option flows. That's when charm doesn't work. One of the things that we're making as we evolve here in Volland is a lookout tower. So, stuff like... how are high yield bonds doing, how are Treasuries doing? If yield goes way up in treasuries most of the time that's a drag on equities because there is an alternative to equities in returns. So, I mean, there's things like that that we are going to have like a lookout tower of other things that will work other than charm. So essentially the answer is when somebody else has an incentive to trade, they may influence a market more than 0DTE trader.
Question: So the other day, yesterday, actually on the daytrading chat, you came in and you talked about when vanna is low, that can mean the market can have wilder swings. Is that because there's lower liquidity in the market? So therefore, because there's more liquidity, obviously there are wild, wild swings and whatnot. Can you explain that again a bit more to me? Because that became interesting to me when you guys were talking about it.
Answer: So when I said low liquidity in the market, I think somebody out there sent me a note that they pointed out that they prove there's liquidity in the market or whatever. What I'm saying, when Vanna is low under pretty like 50 million, you know, like when it's really, really low, that means that option market makers are getting wider spreads... and wider spreads between the bid and the ask is a sign of liquidity for the option market makers. Now if there is not a lot of vanna in Volland, like there's not a lot of supportive flows, like there's not a lot of puts being bought or something like that... To me that's a sign that (then, you know, this is kind of supported by what other people have told me too) people aren't buying puts. That means they're not hedging, and if they're not hedging, that either means they have a ton of dry powder, which... Why would you do that right now when markets are going up, OR they are losing liquidity. And with higher interest rates, even on the overnight RRP, with higher interest rates overall, it would not surprise me if liquidity is going down. And they keep you know, you see what treasuries are doing right now. Interest rates are going up. So we noticed, and you can look on the studies page, we noticed that... as aggregate vanna comes down, you begin to see wilder swings in the market. And I think that's because you are going to see a lot less liquidity. When Vanna comes down close to zero and those bid ask spreads are getting wide, that's when I think option market makers have less liquidity and then, you know, all hell breaks loose. I think that's the situation we were in during COVID and it was also the situation we were in in Volmageddon in February 2018. So, that's kind of what I was getting at. It's also why I said in my tweet, this is why we're not at that point yet. We're not at a point yet where liquidity is a problem yet, but it is a sign that liquidity is coming down from investors' pockets.
Question: A lot of us know when we're trading during the day, we often will look at charm. Sometimes I know that that charm will change rapidly. When we're looking at charm, and from your explanation with gamma, should we also combine our charm readings with what's happening with gamma, or should we just stick with the charm and vanna? If we're looking for like, swings, these sort of like entries or whatnot, because normally, especially with swings and all that, you see that gamma is what we should look at for our entries and supports and resistances. But what about intraday? Should we use another metric as well to kind of give us more confluence?
Answer: When it comes to charm, yes, gamma has an impact intraday. However, it is easy to read through when you're looking at charm. You know that when it is red charm, below the money, you know that those are customer-bought puts and dealers are short puts. And so you already know that that big red bar is charm underneath. It is also going to be big negative gamma. So the other thing too is that Gamma doesn't show you the full picture as well either. Let's say you have a big red bar underneath, but it's just because you pass over a certain call level... dealers were short calls, the market advanced, and now that call is out of the money so now it's red. The dealers are really more impacted by the reduction in implied volatility at the end of the day, since it's going to zero. And a lot of times it's like 20% at 1:00 p.m. So I mean, they're more impacted by the vanna and charm impacts than they are in gamma toward the end of the day. In the beginning of the day, I would look at gamma to see when dealers will, for lack of a better word, "freak out". Like when they say, "Uh oh, we're way offsides here. The market has moved too much for our position. We have to hedge now." I think that's a good time to use gamma. It is something that I even would have to master. When we made Volland Live available, I only had it for about a month longer than you. So I obviously know the theory behind it and the study behind it, but there's part of it that's an art. I think that when you see the big gamma numbers on the cumulative gamma, on the intraday, I think that's probably a good study to do when, I'm thinking about it now as I'm talking through it. It's probably a good study to do like "What is that gamma threshold range when dealers will step in to defend their position, I might say, on a Bank of America paradigm?" or something like that. So in short, I think the answer is at the beginning of the day, and I'd have to probably do a study to determine what the gamma thresholds would be in order for dealers to hedge at some point.
Question: I was just wondering if we should also have other metrics up there as well, like where should we look at like vanna as well? Or is it just that we should only look at a charm because charm just tells us everything else that we know?
Answer: Vanna would show you the same exact thing that charm is, except it would just be showing you probably like opposite signs. But yeah, if you're doing intraday vanna, it would just show you the same exact thing that charm is showing you. Gamma would be a little different though, and mainly because you don't know if it's in the money long calls or in the money short puts. They would show the same in charm, but they wouldn't show the same in gamma. The more and more I think about it, the more I think you might have a point. When I say lines in the sand particularly in like the BofA paradigm, I'm wondering if the gamma levels are what ignites the hedge and not the charm levels. I mean the charm level's line in the sand works pretty well. I bet that it would line up interestingly with cumulative gamma levels that we would see. That's interesting. It's an interesting thought; very curious.
Question: How could one use DAG as a supplemental metric to charm for intraday trading?
Answer: I would use gamma and see that as support and resistance as opposed to DAG. The reason why is because basically the only thing that DAG is doing that is changing anything with the calculation of gamma is flipping the sign of anything above current price from negative to positive or vice versa. And the reason why is we were trying to display, particularly on cumulative, we were trying to display a bullish versus bearish indicator. If you were to do it on intraday trading, first of all, gamma is always strongest at the money. And if you were to cross over the large DAG strike, like let's say it's 4500 and you go from 4501 down to 4499 and it's only a two point decrease. It really doesn't make that much of a difference. But in DAG, we would show like, you know, 2 billion flipped from positive to negative and that might throw you off a little bit.
Question: I have a question that adds on to the last question. You kind of mentioned that say, there's a big gamma strike at 5500 and you move over it by two points. How do you know that it's now going to act as a resistance instead of a support level?
Answer: When you're talking about something like resistance and support as it relates to gamma, more often than not, when you read about it in like SG kind of things and they say, "Oh, it's a big gamma strike. This is going to act as resistance or it's a call wall or something like that." That is more of a vanna phenomenon. It is not that much of a gamma phenomenon. Because gamma doesn't flip sign as you cross over it - vanna does. And so, that changes dealer behavior. And that is what you're looking for when you're looking for support and resistance, is a change of dealer behavior. Now, when you see a strong positive gamma strike, it will provide a little bit of resistance as you approach it. And once you cross over it, it's not all of a sudden like a "breakout". It will still be resistance as you start going, because it is just becoming less and less of a resistance as you keep going. So when I say it's like support and resistance, I'm more or less talking like generally like strong positive gamma strikes are like support and resistance, but if you cross over it, you're not going to break out. Like it's not like a technical indicator where that happens. That's why I personally even like to look at regular gamma. As you cross over that strike, it is still impacting price negatively as you're going away from it. But it's less impactful as you pass it. Because gamma is not changing sign, so it's having the same impact... it's almost like "How much of an impact is it having?"
Question: Just to clarify, when that strike crosses that gamma area, we don't necessarily look for price to react to other entities to see how price is, so once that area is crossed, we know.
Answer: Not because of Gamma, because of vanna, you might expect it to stop. I know that online, all the "GEX addicts" would have you believe that because of gamma, we are stopping at this strike. It's not true. It's because of vanna. There was nothing at the strike. I mean I always like to look at charm myself and Vanna too, because then you know how strong that really is. If you're looking at Volland Gamma, it's obviously more accurate than just OI-based gamma. If you're talking about intraday, I mean know, again, I really want to study the impacts of cumulative gamma on if that is the threshold that dealers are looking at. So if I look at cumulative gamma, I could see like, "Oh, this is when they're defending this position." And they tend to be long calls, too. So it makes sense that upside lines in the sand are defended so much easier. Seems like downside is not as defended as easy. But you never know, I guess. To me, charm levels kind of encapsulate everything that I'm looking for. I was just kind of more or less talking from a data standpoint, why gamma doesn't necessarily cause a breakout once you cross that line. The stopping power, once you cross those lines, are from vanna and not from gamma. Vanna and charm are going to show you the same thing with the opposite sign.
Question: What's your take on the vega?
Answer: Essentially vega is looking at impacts of implied volatility from a PnL standpoint and not a delta standpoint. Vanna can also be interpreted as the movement in vega as it relates to the price in the underlying. However, dealers are not hedging PnL. It is good for you to pay attention to Vega when you're putting on an option trade, but when we're talking about dealer hedging, they are neutralizing their PnL through their share purchases. Their share purchases are determined by delta changes. So that's why we in Volland focus on gamma, vanna, and charm... because gamma, vanna, and charm focus on the changes in the deltas, not the changes in the PnLs.
Question: Does the concept of "max pain" have any usefulness for intraday trading? Can one look for confluence between it and charm?
Answer: Somebody once defined for me what "max pain" meant. And that meant, like, most people are losing buying and selling options. And that's where price trends towards. I think there's a confluence in the fact that sometimes large OI strikes also become large vanna strikes. But there's no flow reason why the max pain thesis would be true. The max pain thesis is a data artifact of the whole vanna thesis. So I think, when you say "can one look for confluence between it and charm," the answer is - just look at charm and the max pain thing will manifest itself. The reason why I would say look at charm and max pain would manifest itself as opposed to the other way around is because OI is two-sided. There are buyers and sellers that have those open contracts, and you don't know how strong dealers are on one side or the other. You probably noticed there are times... like anything with a multiple of 25 in SPX as the largest OI... but when you go into Volland and you look at charm, sometimes 4510 has the largest charm, sometimes 4530 has the largest charm. And that's because dealers are more one-sided on that strike than they are in the 4500 strike. So a lot of times, when people are looking at that max pain thing, they're just looking at OI and saying, "This is where most people lose."
Question: A couple of days ago, the 4500 gamma strike was the largest positive strike. What I have found is that the charm effect can play out multiple times off of large positive gamma strikes. My question is whether this effect actually is caused by vanna.
Answer: Vanna/charm on the intraday. Yes, the whole playing off of it... Because the whole the whole thing is, is that charm and vanna reverse when you cross over the strike. Because it crosses over, that's when the implied volatility changes changed the behavior of the dealers. And on 0DTE, that is more profound, because implied volatility is going in one direction, which is down. Implied volatility will be zero at the end of the day on 0DTE options. And because of that, if you cross over that strike and all of a sudden implied volatility increases, then the dealers will be doing the same thing. So two things have to happen in order for that strike to have the difference. In 0DTE that's not the case. Implied volatility will become zero. So, you know the direction it's going in. So you know that crossing over, you know, I guess you could say negative charm or positive vanna strikes, is going to change the dealer behavior when you cross over them. And that's why I think our 0DTE idea works so well is because implied volatility is going in one direction, time is going in one direction. So, you know what those people have to do.